Overview of Research interests in Software Engineering

My software engineering research program is primarily focused in two major areas - Software Quality and Software Applications in Biomedical Devices.

In software quality, I am developing a streamlined software development strategy which includes both a development process and quality management process which is easy to learn and to implement. The goal is to offer students, and small businesses a means for ensuring just-enough project planning, configuration management, and quaity control to be possible for small-scale projects. I am also developing a testing strategy for software used in research and a set of automated testing tools which support use of the strategy in Matlab.

My interests in software applications are varied. Currently I am developing software tools for automated signal quality analysis of biosignals. I am also developing a modelling tool that can be used to reverse engineer EMG recordings in order to estimate physiolgoical parameters which drive signal generation. Also, I am involved in the development of an EEG monitoring system with context-awareness, which can react appropriately to user states.

Projects of Interest

CleanEMG - Automated Software Quality of sEMG, MScE Candidate , Yiyang Shi

CleanEMG is an ongoing research project that aims to produce automated solutions for analysing signal quality in surface electromyography data.  It is part of a breaoder research program looking to provide clinicians and clients with  reliable electromyography-based assessment and rehabilitation tools, and/or assistive devices. The CleanEMG project is shared bewteen researchers at UNB and our colleagues at Carleton University.

• GD Fraser, ADC Chan, JR Green, DT MacIsaac, "Automated Biosignal Quality Analysis for Electromyography using a One-class Support Vector Machine," IEEE Transactions on Instrumentation and Measureiment, 63(12), 2014.

FRIEND - Brain Monitoring for Adaptive Assistive Systems, PhD Candidate , A Morris

Brain monitoring to infer human conditions such as attention and arousal are useful in both rehabilitative and non-rehabilitative applications. This project investigates agent-based architectures which support non-rehabilitative devices in adapting to inferred user states through brain monitoring. An example would be a car which sounds an alarm when a driver becomes drowsy.

.

Ontology-based Unit Test Generation- MCS Graduate, Valeh Hossienzadeh Nasser

Providing a tool for test experts to define custom coverage criteria can potentially increase quality of automated test suites for software testing. Knowledge engineering technniques can be used to offer this control to test experts using automated test case generation technologies. The ontology-based method used in this work facilitates the enrichment of test oracles with test expert knowledge about error-prone aspects of software under test and allows for custom coverage criteria definition.

• VH Nasser, Weichang Du, Dawn MacIsaac, "Knowledge-based Software Test Generation", The 21st International Confernce on Software Engineering and Knowledge Engineering, Boston USA, 2009.